Pancreatic ductal adenocarcinoma is the 4th leading cause of cancer death in the United States and it has the lowest survival rate for any solid cancer. The goal of personalized cancer therapy is to identify molecular defects that predict drug responses or to test the cancers themselves for drug sensitivities and treat the patient with the optimal drug regimen. The pancreatic cancer genome project identified heterogeneity in the molecular alterations of pancreatic cancers indicating the need for personalized cancer therapy. To this end, we have identified molecular predictors of pancreatic cancer chemotherapy responses. We know that the majority of genes mutated in patients with familial pancreatic cancer are in the homologous DNA repair including BRCA2/BRCA1/PALB2/Fanconi Anemia pathway mutations. Pancreatic and other cancers with defects in homologous DNA repair are exquisitely sensitive to Parp inhibitors and to other drugs that cause double-strand DNA breaks. The Parp inhibitor, olaparib, now in phase 2 clinical trials, has shown a 44% response rate in patients with ovarian cancer. Therefore, we propose a clinical trial randomizing patients to a low-dose DNA strand break regimen of monthly cycles of chemotherapy (irinotecan, cisplatin, bleomycin and mitomycin C) with or without the oral Parp inhibitor, olarparib. We will target patients most likely to respond to this protocol including patients with known BRCA mutations, familial pancreatic cancer, or Ashkenazi ancestry as well as patients with usual pancreatic cancer. Patients and their cancer samples will undergo gene testing to identify gene defects in homologous DNA repair to determine if gene testing can identify patients most likely to respond to Parp inhibitor based therapy. Cell lines will be developed from patient samples to correlate clinical responses to therapy with responses in the laboratory. Circulating mutant DNA will be measured to determine how well it predicts response to therapy. PUBLIC HEALTH RELEVANCE: Pancreatic cancer is the deadliest of the common cancers and the 4th commonest cause of cancer death. Clinical trials that utilize personalized cancer therapy are needed for this disease. We propose a randomized phase 2 trial using a very promising therapy, the oral Parp inhibitor, olarparib, combined with therapies for patients with pancreatic cancer predicted to respond to these drugs based on their genetic profiles and in vitro responses to these agents.